À¶Ý®ÊÓƵ

°Õ¾±³Ù±ô±ð:ÌýStochastic subgradient method converges on tame functions

Abstract: In this talk, I will present a paper () on the convergence of stochastic subgradient method in a nonconvex and nonsmooth setting.

Title: How to Escape Saddle Points Efficiently

Abstract: We will discuss the 2017 paper named in the title of the talk, by Jin et al.

Title: Announcing new Reading Group this term on the topic of "Entropy and Counting"

Abstract:

A new Reading Group will be starting this term on the topic of "Entropy and Counting". The talks will take place weekly on Tuesdays from 2:00 p.m. to 3:00 p.m. in MC 6486 starting on Jan 22th.

Title: Introduction to Entropy and Counting

´¡²ú²õ³Ù°ù²¹³¦³Ù:ÌýIn the first talk of our reading group, we introduce the notion of entropy and derive some of its basic properties.

Title: Induced Poset Saturation

Abstract: In Graph Theory, we say that a graph G is H saturated if G contains no copy of H as a subgraph,

Title: Additive Approximation of Chromatic Number in Apex-minor-free Graphs

Abstract: In this talk we will see a t+3 additive approximation to the chromatic number for t-apex-minor-free

Title: Convex Optimization Problems in Domain-Driven Form

Abstract:

In this talk, we introduce the Domain-Driven form for convex optimization problems and show how general it is by several examples; including LP, SDP, geometric and entropy programming, and quantum entropy optimization.

Title: Applications of the entropy method

Abstract: In the second talk of our reading group on entropy and counting.

Title: Fractional coloring with local demands

Abstract: In a fractional coloring, vertices of a graph are assigned subsets of the $[0, 1]$-interval such that adjacent vertices receive disjoint subsets. 

Title: Quantum Walks on Oriented Graphs

Abstract: A quantum walk on a graph is defined based on a Hermitian matrix associated with the graph, such as the adjacency matrix.